Hello Guest it is March 29, 2024, 11:31:09 AM

Show Posts

This section allows you to view all posts made by this member. Note that you can only see posts made in areas you currently have access to.


Messages - mc

Pages: « 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 »
311
General Mach Discussion / Re: lathe threading pitch error
« on: January 20, 2013, 09:34:36 AM »
If you're confident the Z-axis is accurate, the only thing that can be of is the spindle speed.

How confident are you that the spindle speed is correct?

312
General Mach Discussion / Re: Lathe tool change problem
« on: January 15, 2013, 11:26:24 AM »
xml is attached.

I had a few PM's with Hood last night, thinking I was doing something stupid in the g-code, until I remembered about this problem.
Last time I noticed it because I heard the stepper stall as I moved into position before starting the spindle, whereas this time I had the spindle running before the move, so I never heard/noticed the stepper stalling.
I know that to get around the problem, I can just add a move to the aproximate position I need for the next tool, but I'd like to solve the problem than continually have to work around it.

Here's a snippet of g-code that caused the problem last night
Code: [Select]
N270 M5
N280 G0 X125 Z100
N290 T0404 (ER Tool gang with 6.8mm drill in rear)
N300 M3
N310 ;M830 X0 Y0 Z-33.5 R2 Q3 W0.5 P0 D0 F100 I0 J0 L1
N320 M830 P310
N330 M5
N340 G0 Z10
N350 T0606 (2mm MDT)
N360 M3
N370 G0 X16
N380 G0 Z-31
N390 F50
N392 G1 X14.8
N394 G1 Z-32 X14
N400 G1 X0
N410 G0 X16
N420 M5
N430 G0 Z0
N440 M30

Line 340 to 430 were what I added last night to make use of a MDT tool to chamfer and part off.
Line 370 is where the problem occurs, because after the tool change, the x-axis has to move near full travel (about 80mm travel) to reach X16.
If T6 was commanded to X16, it would have to move 8mm (I run in diameter mode), so it's as if Mach tries to move that 80mm in the time it would take to move 8mm.
(X0 for T4 is just of home position, whereas X0 for T6 is nearly full opposite travel)

The DROs are updated to the new tool offsets as soon as the toolchange is called, so Mach, or at least the SS, knows where the tool is, but on the next move, Mach fails to allow extra time for any additional movement caused by the changed tool offsets.

The strange thing is, this problem never affects the Z-axis. I've got tool changes where the Z-offset is 100mm different, yet on the next move after a tool change, all the moves are spot on.

I'm currently running the latest stable SS plugin (15ogx I think), but not sure what version of Mach I have on the lathe computer.
Once I get this latest batch of parts run, I was going to try updating to the latest recommended versions.

313
General Mach Discussion / Lathe tool change problem
« on: January 14, 2013, 05:32:10 PM »
This problem has existed since I started using my lathe, but I managed to tweak the g-code to get around it, but having just hit the problem again tonight, I want to figure what is actually causing the problem.

When changing between tools in the g-code, if there is a large distance on the X-axis needing moved, lots of steps get lost/the stepper stalls.
Tool changes are fine as long as the x-axis distance needing moved after the tool change is minimal (or is combined with a long Z move).

For example, if the new tool offset is 80mm different, and needs a physical movement of say 2mm to get to the next commanded position things are fine, or if it needs to move 100mm in the Z and 80mm in the X, things are fine.
However, if the physical movement needed is say 50mm in the X alone (or combined with a few mm in the Z), the x-axis will stall.

It's as though Mach ignores the max speed of the X-axis for that first move after a tool change.
If the first move is combined with a long Z move, because the two are interpolated to move diagonally, things go fine.

I'm not sure if it's related to some setting (G0 moves never cause any problems, and I've even tried reducing the X-axis max speed when I first had the problem), a bug with the SmoothStepper, or something wrong with Mach.

Anybody got any suggestions?

314
Show"N"Tell ( Your Machines) / Re: Matchmaker Revival
« on: January 14, 2013, 11:49:49 AM »
Some carberetor cleaner worked wonders on the metering units, as it's pretty good at disolving the remains of oil/varnish. The ones that couldn't be unblocked were all full of black solidified oil, which given the size of orifices, would be pretty much impossible to unblock without dismantling them. Boiling them up in some detergent might work, but they're not that expensive to replace, plus the only way to get to the quill one is with the head off, and I ain't risking having to take it back of for the sake of trying to save a few pounds!

I filled the auto lube up with ATF last night and left it running, and there's a coating of red in most places where oil should be, although there's at least one pipe leaking. I noticed it was pretty badly pitted when I connected everything back up, but had hoped it wasn't that bad. I just hope I can get the other end off and back on without having to remove the table.

The main pipe was pretty badly corroded, I'm guessing caused by the drain hole being blocked, so it's probably been sitting under coolant/water for a while. I think the final nail was the ballscrew spring cover hasn't been sitting in it's housing right so when I moved the table to full travel it's been pushed down and caught the pipe. The ways aren't exactly in pristine condition, but they seem clean enough, and only obvious damage are a few scrapes most likely from bits of swarf that have been caught over the years.

Those servos are tempting, but I'm skint until the end of the month!

315
Show"N"Tell ( Your Machines) / Matchmaker Revival
« on: January 13, 2013, 06:52:36 PM »
First off, I'm blaming Hood for this. If it wasn't for him pointing me in the direction of this mill, I would of been happily on my way to building a nice desktop sized router!

Anyway, after Hood's help, a few months ago I became the owner of a Matchmaker mill. The actual knee mill is a Shizouka ST-N, with a Summit aka One Arm Bandit tool changer.


Shizouka ST-N by mc_mtb, on Flickr

After removing the head, and with some questionable use of a forklift and some skates, it made it into the workshop.

In Position by mc_mtb, on Flickr

I'm not in a great rush to get this machine up and running as I've got various other projects/jobs that have priority, but inbetween swapping bits in my lathe, I've been tracing wires, doing some cleaning, and getting the autolube up and running. After cleaning out the autolube unit, I filled it up and left it running overnight to see if any oil appeared on the ways. And despite managing to pump half a litre out the resevoir, nothing appeared anywhere it should, or even on the floor. The oil was finally located sitting in a recess under the table, which had a blocked drain hole, and was getting there because of a burst alloy pipe.
While having the ballscrew support of the end of the table, I stripped and cleaned the distribution block, and found two of the Bijur metering units are blocked, so I stripped the other distribution blocks, and found one more blocked on the knee, along with the main quill one. Buying some new metering blocks is on this weeks list of jobs.

Here's the distribution block refitted with a new copper pipe -

Autolube Distribution by mc_mtb, on Flickr

As it stands, my plan is to use a Kflop with Kanalog running through Mach for the main control, with high voltage steppers and drivers for the axis running closed loop through the Kflop.
I'd like servos, but I can't justify the cost at the moment. 3 steppers and drivers are roughly the same cost as a servo for a single axis, so I'll live with the slowness of steppers for now, however by using the Kflop upgrading to servos later will be a pretty straightforward job.

316
General Mach Discussion / Re: spindle load feedback to mach3
« on: December 17, 2012, 05:02:09 PM »
I'm not sure if I've got the manual for the same VFD, but according to this manual (http://www.vacon.com/ImageVaultFiles/id_2862/cf_2/Vacon-10-User-Manual-DPD00717C2-EN.PDF) all the information you want is available via Modbus.

317
General Mach Discussion / Re: spindle load feedback to mach3
« on: December 17, 2012, 12:03:26 PM »
What inverter is it?

It can be done, however without knowing what inverter you have, nobody can really tell you.
Provided the inverter can output the required information via RS485 (aka Modbus), and you have an adapter to connect RS485 to your computer, it is doable.

318
It means I can push the steppers a little bit harder if needed and know if I've pushed them too far.
But also with having the encoder wiring already in place, combined with high voltage stepper drivers, a servo upgrade should be a quick and relatively painless move.

319
Ok I see what you are meaning now, you are using both tool calls and G52's in the same code dependant on which tools are being used.
Best way is to just set the tool table up and forget the G52's

Hood

Will that not mean the lathe will stop and wait for a toolchange between drills?

320
Very true.

Yeah. My current plan is Kflop+Kanalog, as that allows me to close the loop with steppers and have differential encoders, then allow an easy upgrade to servos later. However got a couple upgrades I want to do to the lathe first, so the mill is pretty low down the priority list.

Pages: « 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 »